Introduction

This study was performed to investigate the failure mechanism of one specific hip arthroplasty cup design that has shown a high clinical failure rate. The aim of this study was to identify general design problems of this polyethylene inlay.

Background:

Total ankle replacements (TAR) are not as successful as total hip or total knee replacements. A three-time increased revision rate is reported in registry data [1]. Therefrom, wear associated revisions are frequent [2]. However, there is little knowledge about the wear behavior of TAR. This may be partly related to the fact, that currently no standard for wear testing of TAR exists.

The aim of this study is to define a biomechanical valid, force-controlled test specification for level walking of TAR.

Background:

There is little knowledge about wear performance of total ankle arthroplasties (TAR). However, revisions rates are high for TAR [1] and wear associated revisions are frequent [2].

Therefore, the aim of this study is (1)

To test the wear behavior of a TAR using a biomechanically valid testing scenario.

Introduction

Concerning biomechanical research, human specimens are preferred to achieve conditions that are close to the clinical situation. On the other hand, synthetic femurs are used for biomechanical testing instead of fresh-frozen human femurs, to create standardized and comparable conditions. A new generation of synthetic femurs is currently available aiming to substitute the validated traditional one. Structural femoral properties of the new generation have already been validated, yet a biomechanical validation is missing.

The aim of our study was to analyse potential differences in the biomechanical behaviour of two different synthetic femoral designs by measuring the primary rotational stability of a cementless femoral hip stem.

Introduction

Osteolysis and aseptic loosening in total hip replacement (THR) is often associated with polyethylene (PE) wear. This caused interest in alternative bearing surfaces. Since the mid nineties, research focused on hard-hard bearings like metal-on-metal (MOM) or ceramic-on-ceramic (COC). However, concerns remain about biological reactions to metallic wear debris or failure of the ceramic components. A new approach to reduce wear with a minimized risk of failure may be the use of a metallic cup in combination with a ceramic head, the so called ceramic-on-metal bearing (COM). The aim of this study was to estimate the wear behaviour at an early stage of this COM bearing type in comparison to COC bearings using a hip simulator.

Knee wear simulator studies are performed to evaluate wear behavior of implants.

Simulation of the human gait cycle is often carried out continuously, without considering resting periods as they are part of patient’s daily live. In addition to dynamic activities like walking, daily activities also consist of static periods like standing, sitting or lying. During the day dynamic activities alternate continuously with static periods and most of the day is spent in passive periods, where no joint motion occurs. Such resting periods have not yet been considered in prosthetic knee wear tests. Implementing resting periods may cause an increase in friction and thus increased wear of the implant. The aim of the current study was to determine if the implementation of resting periods would increase polyethylene (PE) wear in total knee replacement (TKR).

Two wear studies were conducted using a force controlled AMTI knee simulator on a conventional bicondylar TKR. For the first study, simulation was carried out continuously according to ISO 14243-1. For the second test, four active gait cycles according to ISO 14243-1 were followed by one resting period cycle. In both tests 5x10E6 active load cycles at a frequency of 1 Hz (resulting in additional 1.25x10E6 pause cycles for the second test) were applied. Wear was measured gravimetrically and wear scars were documented photographically.

The mean wear rates measured 2.85 ± 0.27 mg/10E6 cycles for the ISO test without considering resting periods and 2.27 ± 0.23 mg/10E6 cycles for the test with resting periods implemented. There was no significant difference (p=0.22) in wear rate between both tests.

The inserts showed similar wear scars in both tests and no relevant differences in dimension and localization on the surface. Therefore the wear behavior after the two tests was similar.

Since wear is one of the most limiting factors for implant longevity, proper preclinical wear studies are essential. Based on the results of this experimental wear study, a continuous simulation without additional resting periods seems to be valid in wear simulation of TKR.

Introduction: During hip stem revisions osteotomies allow to remove well-fixed components. Once removal has been done, cerclage wires should secure the osteotomy and support primary stability of the new stem. Stability is important for a bony ingrowth and therefore the longevity of a cementless revision stem.

Tension wires seem to dominate revision surgery and studies only refer to the advantages of cable wires in general. This in-vitro study analyzed the infiuence of both, tension and cable wires on primary stability of cementless revision stems. We aimed to examine the effectivity of wiring a femoral osteotomy, differences achieved with each method, and whether one wire has advantages regarding the fixation concepts of revision stems (meta- and diaphyseal).

Methods: We studied a Ti-tension- and a CoCrWNi-cable-wire. The Helios-stem stood for the meta- and the Wagner-SL-stem for the diaphyseal fixation concept. Each stem was implanted into 3 synthetic femurs and a standardized extended proximal femoral osteotomy was performed. Spatial movements of bones and stems at several sites were explored under axial torques using a high-resolution measuring device. Movement graphs subjected to the sites defined relative movements RM = ΔαZ/TZ [mdeg/Nm]. The osteotomies were locked consecutively with both wires and all compounds were measured again. Wiring was done by a proximal figure 8 and a diaphyseal circular loop.

Results: Compared to the unlocked osteotomy the tension as well as the cable wires caused a changed RM for the stems (p=0.03). Both wires affect an increased stability within the proximal main fixation area of the Helios. Even for the Wagner-SL, usually fixating diaphyseally, a proximal fixation was reached with both wires. A significantly better stabilization could be observed for the Helios using cable wires (p=0.04). The overall RM reached with tension and cable wires was 16.6 and 11.1 mdeg/Nm. The Wagner-SL^® showed no difference in stability between tension and cable wire treatment (p=0.29).

Discussion: Both, the tension and the cable wires support the revision stems in bridging the artificial defect of an extended proximal femoral osteotomy. Especially for the proximal fixating stem, RMs could largely be reduced, while cable wires seem to be advantageous. Preventing a circular constriction leading to an osseous malnutrition, the use of cable wires, however, should be impeded with regard to diaphyseal fixating stems and proximal osteotomies. Comparable results with both wires were reached and none of the wires showed any advantage in this situation. In conclusion, the wires should be chosen depending on the fixation concept of the revision stem.

Introduction: Postoperative periprosthetic fractures are difficult to investigate clinically in scientifically and statistically valid study samples because they are uncommon. However, the combination of advanced age and poor bone quality has been associated with these fractures in conjunction with cementless hip stems. So far, these speculations have neither been supported by clinical evidence nor been investigated experimentally. The purpose of the present study was to analyze in-vitro if the treatment with cementless hip stems increases the risk of suffering a periprosthetic fracture for older patients compared to younger patients. Regarding this manner, we aimed to clarify parameters which possibly can be used preoperatively to assess age related fracture risk and whether the femoral bone quality really plays a role in fracture development.

Methods: An established biomechanical setup was used to provide an investigation on 16 femoral specimens of different age. Prior, the BMDs were measured in 5 ROIs and a cementless hip stem was implanted into each femur. The load bearing of “normal walking” was applied quasistatically under standardized conditions until the fracture occurred. The specimens were arranged by age in ascending order to divide them in the middle. A group of under septuagenarians (< 70y) (mean: 62y) and a group representing an elderly population (≥77y) (mean: 79y) resulted. Important donor data such as body height and bodyweight were considered in the statistical analysis.

Results: The elderly specimens fractured at significantly lower forces (< 70y: Fmax=5,308N; ≥77y: Fmax=2,519N; p< 0.01). Pearson’s test revealed a correlation for Fmax [N] and age (p < 0.01; r = −0.64); and for Fmax [%BW] and age (p < 0.01; r = −0.69). Fracture loads were found to correlate strongly with age (p=0.01), all used ROIs (e.g. for Ward’s triangle: p< 0.01) and BMI (p=0.04). Decreasing CCD angles were found with increasing age (p < 0.01).

Discussion: In patients with advanced age treated with cementless hip stems the risk of suffering a periprosthetic fracture is significantly higher. The identification of specific fracture development variables in geriatric populations can be extended to a preoperative check list to aid clinicians in practicing effective risk assessment. Criteria such as BMD, BMI and CCD angle should be included: A fracture risk remains in patients around 80 years of age or older, with a Ward’s triangle BMD below a value of 0.500g/cm2, or a BMI > 33kg/m2. Depending on patient activity, one single factor should not be viewed as an exclusion criterion for a cementless hip stem, whereas the cumulation of them should alert the surgeon.

Introduction: In recent retrieval studies over-penetration of cement, incomplete seating of the prosthesis with a resultant polar cement mass, or both, have been associated with femoral failures of current generation resurfacing arthroplasties. We developed a laboratory model to analyze differences in cement penetration, cement pressures and interface temperatures for hip resurfacing arthroplasty.

Materials and Methods: A carbon foam was demonstrated to closely simulate human femoral heads. Custom aluminum shells were made by DePuy with the same inner geometry as the femoral resurfacing components. (ASR™ system, Size 49, DePuy; Leeds, England).

Analyses of six different cementing techniques (cemtech) were performed using high viscosity (HVC) (Smart Set GHV, DePuy, Blackpool, England) and low viscosity cement (LVC) (Endurance, DePuy, Blackpool, England):

Manual application HVC

A force of 150N was used to press five shells in each cemtech group on foam specimens. During seating cement pressures and polymerization heat 5 mm under the foam surface were measured.

Specimens were cut into quarters, surfaces were digitalized and cement penetration areas and depths were quantified using a pixel-analysis-software. The effects of the cemtech were examined by Kruscal-Wallis and Mann-Whitney-U-tests (two-sided, p-value< 0.05, SPSS)

Results: The mean cement pressures increased going from cemtech A to E. HVC cemtech C and E showed higher pressures than the comparable LVC cemtech B and D.

Maximum temperatures were A) 36.0± 4.1°C, B) 45.0±5.7°C, C) 36.2±4.2°C, D) 53.5±2.5°C, E) 48.3±6.5°C and F) 53.2±12.6°C. D, E and F exceeded 50°C.

A provided even cement penetration over the available fixation area without involvement of the internal area and the stem. Cemtech that used LVC cement (B, D and F) showed higher interior area cement contents than HVC (A, E and C). The cement content in the interior area was A) 39.3±26.4mm2, B) 72.1±16.9mm2, C) 37.7±10.5mm2, D) 99.0±24.6mm2, E) 67.5±15.6mm2 and F) 121.0±29.0mm2.

A showed mainly complete seating with a cement mantle thickness of 0.5±0.7 mm. All other cemtech had incomplete seating in all specimens with significantly thicker polar cement mantles (p=0.032) up to a maximum of 4.6±1.2mm for E.

Discussion: Component filling cemtech and LVC resulted in variable degrees of over-penetration, exposure to high temperatures or a risk for incomplete seating, which have been associated with bone necrosis and early fracture. The use of the manual application and HVC cement showed clear advantages in our model. It was possible to utilize all of the available fixation area without negative effects.

Resurfacing hip implants differ in macro- and microstructure. Manufacturing related parameters like clearance or carbon content influence the wear behaviour of these metal-on-metal bearings. The aim of this study was to analyse the main macro- and micro-structural differences of commercially available resurfacing hip implants.

Ten different commercially available resurfacing hip implant designs were included in this investigation:

- BHR^® (Smith& Nephew/MMT)

The heads and cups were measured in a coordinate measuring machine and radial clearance as well as sphericity deviation were calculated. Surface roughness measurements were carried out. The microstructures of the heads and cups were inspected using SEM and element analysis was performed using EDX to identify carbides and the alloy composition.

The mean radial clearance was found to be 85.53 μm. The range was from 49.47 μm (DePuy, ASR^®) to 120.93 μm (Biomet, ReCap^®). All implants showed a sphericity deviation of less than 10 μm. The highest sphericity deviation was found to be 7.3 μm (Corin Cormet^® head), while the lowest was 0.8 μm (Smith& Nephew BHR^® head). On average, the heads tended to have a higher sphericity deviation (4.1 μm, SD: 2.3 μm) compared to the cups (2.7 μm, SD: 1.4 μm). SEM revealed that most manufacturers use a high carbon alloy casting manufacturing process combined with heat treatment after casting (Corin Cormet^® and Wright Conserve^®: head and cup; DePuy ASR^®: cup; Eska BS^®: head).

Commercially available resurfacing hip implants differ in design and manufacturing parameters, including macro- and microstructure, which are critical in achieving low wear and ion release. This study was designed to aid in the understanding of clinical observations. Also, specific information is now available for surgeons choosing an implant designs.

Resurfacing hip implants differ in macro-and microstructure. Manufacturing related parameters like clearance or carbon content influence the wear behaviour of these metal-on-metal bearings. The aim of this study was to analyse the main macro- and micro-structural differences of commercially available resurfacing hip implants. Ten different commercially available resurfacing hip implant designs were included in this investigation:

BHR^® (Smith& Nephew/MMT) Durom® (Zimmer)

The heads and cups were measured in a coordinate measuring machine and radial clearance as well as sphericity deviation were calculated. Surface roughness measurements were carried out. The microstructures of the heads and cups were inspected using SEM and element analysis was performed using EDX to identify carbides and the alloy composition. The mean radial clearance was found to be 85.53μm. The range was from 49.47μm (DePuy, ASR®) to 120.93μm (Biomet, ReCap®). All implants showed a sphericity deviation of less than 10μm. The highest sphericity deviation was found to be 7.3μm (Corin Cormet® head), while the lowest was 0.8μm (Smith& Nephew BHR® head). On average, the heads tended to have a higher sphericity deviation (4.1μm, SD: 2.3μm) compared to the cups (2.7μm, SD: 1.4μm). SEM revealed that most manufacturers use a high carbon alloy casting manufacturing process combined with heat treatment after casting (Corin Cormet® and Wright Conserve®: head and cup; DePuy ASR®: cup; Eska BS®: head). Commercially available resurfacing hip implants differ in design and manufacturing parameters, including macro- and microstructure, which are critical in achieving low wear and ion release. This study was designed to aid in the understanding of clinical observations. Also, specific information is now available for surgeons choosing an implant designs.

For wear testing of knee implants, ISO 14243 is the most used testing protocol. In force control, this standard requires linear motion restraints for simulation of ligaments. The aim of this study was to investigate if a nonlinear, physiological motion restraint would influence the wear behaviour of the implants.

A wear study was performed on a highly conforming knee implant design. Three implants were tested forced controlled according to ISO 14243-1 on an AMTI knee simulator. Linear motions restrain of 30 N/mm for AP-translation and 0.6 Nm/° for IE-rotation were applied as required per ISO 14243-1. A second wear test was performed on the same implant design. Based on the data given by Kanamori et al. and Fukubayashi et al., a physiological, nonlinear ligament constraint model (sectioned ACL) was adopted and implemented in the simulation. The implants were pre-soaked and a soak controls was used. Wear was measured gravimetrically.

A mean gravimetric wear rate of 2.85 mg/10E6 cycles was found for the implants which were tested using a linear motion restraint as required per ISO 14243-1. Simulating a physiological, nonlinear motion restraint resulted in a 60% increase in gravimetric wear (mean gravimetric wear rate: 4.75 mg/10E6 cycles). As expected, the kinematics of the implants differed between wear tests. The mean AP-translation increased from 2.89 mm (linear motion restraint) to 4.82 mm (physiological motion restraint). A similar behaviour was observed for the IE-rotation. The IE-rotation increased from 4.09° (linear motion restraint) to 5.94° (physiological motion restraint).

The reaction of the ligaments is not linear in the human knee joint. This study showed that wear and kinematics change when simulating physiological ligament reactions. Wear increased by 60%, an effect which can likely be credited to fundamental differences in kinematics. The ACL is commonly sacrificed during surgery. Thus, more attention should be paid to ligament simulation when performing wear tests on knee implants.

The introduction of mobile bearings for unicompartimental knee implants resulted in heightened interest in this implant design in the field of orthopaedics. This study aims to determine the effect of the mobile and fixed design concepts on the wear progression in unicompartmental knee implants using a knee simulator.

An unicompartmental knee implant design, which is available in a fixed and mobile version, was tested using a knee simulator. For the wear test, the medial and lateral compartments were implemented in the simulator. To account for the physiologically higher medial load compared to the lateral compartment, a medially-biased load distribution was implemented. The wear test was performed force controlled according to ISO 14243. Wear was measured gravimetrically separately for the medial and lateral compartments. To evaluate implant kinematics, AP-translation and IE-rotation were measured during the simulation.

Gravimetric wear was higher medially than laterally for both designs. The mean wear rate of the medial mobile compartment was found to be 10.70 mg/10E6 cycles, whereas a mean wear rate of 6.05 mg/10E6cycles was found for the medial compartment of the fixed design. Lateral wear rates, which were about 50% lower than medial wear rates, were found to be 5.38 mg/10E6 cycles in the mobile design and 3.23 mg/10E6 cycles in the lateral design. Wear of the mobile design was higher compared to the fixed design, both medially and laterally. Surprisingly the kinematics of both designs were very similar. A low AP-translation of 2.7 mm in the mobile and 2.4 mm in the fixed designs was documented. High IE-rotations of 6.5° and 6.7° for the mobile and the fixed design, respectively, were observed.

In bicondylar bearing knee designs, reduced wear has been reported for mobile polyethylene inlays. This study showed that the wear behaviour of unicompartmental knee implants differs from bicondylar implants and that the introduction of the mobile concept may lead to increased wear.

In hip joint simulator studies, wear measurement is usually performed gravimetrically. This procedure is reliable for metal-on-polyethylene or ceramic-on-polyethylene bearings, in which relatively high amounts of abrasive wear particles are produced. With modern hard-on-hard bearings, volumetric wear decreases significantly up to 100 to 200-fold. The gravimetric method reaches its detection limit with metal-on-metal bearings and even more so with ceramic-on-ceramic bearings. This study establishes a new method of determining wear in hard-on-hard bearings by measuring the amount of worn particles/ions in the serum of hip simulators.

A wear study on three resurfacing hip implants (BHR^®, Smith& Nephew) was conducted using a hip joint simulator. Prior to the wear study, tests were performed to validate the detection power for high resolution-inductively coupled plasma-mass spectrometry (HR-ICP-MS). More importantly the system’s accuracy was compared to the gravimetric method, which is described in ISO 14243-2. The simulator was altered to run completely metal ion free. The ion concentration in the serum was measured every 100 000 cycles up to 1 500 000 cycles and subsequently in intervals of 500 000 cycles using HR-ICP-MS. The implants were neither removed from the simulator nor excessively cleaned during the course of the simulation. Serum was refreshed every 500 000 cycles. The serum samples were digested with purified nitric acid and hydrogen peroxide using a high pressure microwave autoclave in order to measure wear particles as well as dissolved ions. All steps were carried out under clean room conditions. Wear was calculated using the ion concentration and measured serum volume. Wear rates and transition from running-in to steady-state wear phases were calculated.

A detection power better than 0.028 μg/l for Co (cobalt), 0.017 μg/l for Cr (chromium) and 0.040 μg/l for Mo (molybdenum) was found for HR-ICP-MS. The validation of HR-ICP-MS showed good agreement between gravimetric data and measured ion concentrations. The tested implants showed similar wear behaviour. Implant wear resulted in high ion concentrations during the first 380 000 to 920 000 cycles. During this period, a mean wear rate of 6.96 mm3/10E6 cycles was determined. Subsequently, the wear rate significantly decreased to a mean wear rate of 0.37 mm3/10E6 cycles. Thus, a mean ratio between running-in and steady-state wear of 18.8 was found. The mean overall wear volume at the end of the simulation was 4.42 mm3.

This study showed that measuring the ion concentrations in the serum of hip simulators can be used to determine wear in metal-on-metal bearings. The main advantages of this new method are the ability to detect ultra-low wear rates and to precisely specify the duration of different wear phases. Because the implants do not have to be removed from the simulator and aggressive cleaning processes may be skipped, fluctuations in wear detection are extremely low. This in turn leads to a shorter duration of the simulation. Wear rates of the tested implants are low compared to polyethylene. Transferring the results to patient activity, wear would be the same during the first four to six months after implantation as in the next ten years. Minimizing the duration of running-in would be most effective in further reducing wear of metal-on-metal bearings.

Introduction: Aim of the study was to give an overview about the main macro- and microstructure differences of commercially available resurfacing hip implants. The effect of the manufacturing process and the subsequent heat treatment leads to variable microstructures of implant materials. It is undisputable that a low surface roughness and high sphericity improves the wear behaviour. But the radial clearance, the manufacturing process and heat treatment are discussed controversially.

Methods: Resurfacing hip implants with a 46mm head diameter and corresponding cups were analyzed. Commercially available hip resurfacing implants from 10 different manufacturers were included in this investigation. The heads and cups were measured in a coordinate measuring machine (Mahr Multisensor MS 222). A best fit sphere was created from the point clouds and evaluated using analysing software (Imageware 12.1, UGS Corp.). Head and cup radial clearances were measured and sphericity deviation calculated and graphically plotted. Measurements on surface roughness were carried out three times per implant (Mahr Pertometer M2). The microstructures of the heads and cups were inspected by SEM (LEO 440). Surface images were taken using the scanning electron mode. The back scatter electron mode was used to get element weighted images.

Element analysis was performed by EDX (Oxford D. 7060) to identify carbides and the alloy composition. Element distribution maps were taken to separate the single elements.

Results: The mean radial clearance was found to be 85.53μm. The range was from 49.47μm to 120.93μm. We classified all implants into three groups (low, midrange and high clearance). The low clearance group ranged from < 50μm to 75μm, midrange from 75μm to 100μm and high clearance from 100μm to > 125μm.

All implants showed a sphericity deviation less than 10μm. On average the heads tended to have a higher spherical deviation of 4.1μm (SD: 2.3μm) compared to the cups 2.7μm (SD: 1.4μm). Based on the SEM and EDX inspection the manufacturing process, heat treatment and carbide distribution could be clarified.

Discussion: This study gives an overview about the main macro- and microstructure differences of commercial available resurfacing hip implants.

The characteristically unspheric formations of the heads may be due to the cooling process after manufacturing the implant and there is also a relation between the wall thickness of the implant and the unspheric formations. With decreasing wall thickness the implant cools faster locally. Additionally a cup with a thin wall may deform under loading condition and a very tight clearance could be detrimental.

This study will help to understand clinical observations. It still has to be proven that these biomechanical factors influence the clinical performance of hip resurfacing implants.

The aim of the study was to compare mortalitiy and complication rate after operative treatment of pertrochanteric fractures with primary cemented arthroplasty, dynamic hip screw (DHS) or proximal femoral nail (PFN).

283 patients, which were treated betwen 1992 and 2005 for pertrochanteric femoral fractures, except pathologic fractures and a minimum age of sixty years were included. 132 of these 283 patients were treated by primary arthroplasty.

Up to the end of 1999 all unstable fractures were treated by primary total hip replacement. In the year 2000 the PFN was introduced and only patients with severe osteoarthritis and osteoporosis received primary arthroplasty. I possible, more stable fractures were treated with a DHS.

One year mortality was chosen as main indicator as it depends on the surgical trauma as well as the rapid return to preinjury activity and further complications. A one year period was chosen as the mortality ratio approaches that of an age matched reference population after this interval. Influencing cofactors were eliminated by stepwise logistic regression analysis.

It was shown that restoration of the preoperative ambulatory level correlated with survival rate after one year. As elderly patients are often unable to cooperate with partial weight bearing, the primary stability of the device is crucial to allow early mobilisation

Mortality was significantly influenced by age, gender and comorbidities but not by fracture classification. One-year mortality was significantly higher for primary total hip replacement (34.2 %) than for internal fixation (DHS: 18.4 %; PFN 21.4 %) and hemiarthroplasty (13.3 %). Since the PFN and hemiarthroplasty were introduced the over all mortality was reduced from 29 % to 18 %.

Conclusion: For stable fractures a Dynamic Hip Screw (DHS) and for unstable fractures a short proximal femoral nail (PFN) can be recommended. Primary cemented hemiarthroplasty is a viable option for treatment of intertrochanteric fractures if osteoporosis prevents from full weight bearing or if osteoarthritis makes further operations likely. Primary total hip replacement should be avoided, due to the fact that dislocation and mortality were significantly higher than in the other groups.

Introduction: Cement pressurisation is key to achieving good cement-bone interdigitation in THR. To obtain adequate pressurisation the medullary canal must be sealed distally using a cement restrictor. The cement restrictor must remain stable in the femoral canal.

Methods: Five different cement restrictors were evaluated, namely the Exeter Cement Plug, Biostop G, Hardinge, Rex CementStop and a preinjected cement plug. The restrictor was deployed in a sawbone that had been reamed to produce a distal flare, based on radiographic measurements. Low viscosity bone cement pressurised using a cement ram connected to a 10bar air supply. An electronic pressure valve increased the pressure in the cement. Cement pressure and cement restrictor displacement were continuously measured. The pressure valve and recording of measurements was controlled by a customised computer package.

Results: The Rex CementStop withstood the greatest pressures (mean 565.8kPa). This was a significantly greater pressure than any of the other cement restrictors (p< 0.001). Pre-injected cement plugs were able to resist the next highest pressures (mean 350.4kPa). They did not displace but leaked cement and were technically difficult to deliver in the distal femur. Cement restrictors that function well above the isthmus were ineffective (Biostop mean 118.7kPa) or could not be deployed below the isthmus (Exeter). The Hardinge recorded a mean 162.3kPa.

Discussion: During pre-operative templating it is important to consider where the cement restrictor will sit in the femur. When the cement restrictor is going to be deployed beyond the femoral isthmus, an alternate method of cement restriction may need to be used. Universal sized plugs (e.g. Hardinge) function poorly in this situation. Press-fit plugs such as Biostop and Exeter are severely compromised when inserted past the femoral isthmus. Pre-injected cement plugs are variable in efficacy. The expandable Rex CementStop reliably occluded the femur, allowing the highest pressures to be generated.

Aims: To evaluate the function of cement restrictors beyond the femoral isthmus.

Introduction: Pressurisation of cement is key to achieving good cement-bone interdigitation in Total Hip Replacement. During insertion of the femoral stem, pressures of up to 1000kPa may be generated. To maintain pressurisation the medullary canal must be sealed distally using a cement restrictor. As a secondary effect, cement restrictors also prevent excess injection of cement into the medullary canal. To fulfil these functions the cement restrictor must remain stable in the femoral canal.

Methods: Five different cement restrictors were evaluated, namely the Exeter Cement Plug (Stryker, UK), Biostop (De Puy, UK), Hardinge (De Puy, UK), Rex CementStop (A-One-Medical, Netherlands) and a preinjected cement plug (Surgical Simplex, Stryker, UK). The restrictor was deployed in a sawbone that had been rasped to produce a distal flare. Low viscosity bone cement (Surgical Simplex, Stryker, UK) was injected and pressurised using a custom made cement ram connected to a 10bar pressurised air supply. An electronically controlled pressure valve increased the pressure in the cement. Pressure in the cement was measured using a pressure transducer. A linear variable displacement transducer was used to measure movement of the cement restrictor. Leakage of cement around the restrictor was also recorded. Activation of the pressure valve and recording of measurements was controlled by a customised computer package.

Results: The Rex CementStop withstood the greatest pressures (mean 565.8kPa). This was a significantly greater pressure than any of the other cement restrictors (p= 0.027). Pre-injected cement plugs were able to resist the next highest pressures (mean 350.4kPa). They did not displace but leaked cement and were technically difficult to deliver in the distal femur. Cement restrictors that function well above the isthmus were ineffective (Biostop mean 118.7kPa) or could not be deployed below the isthmus (Exeter). The Hardinge cement restrictor recorded a mean 162.3kPa.

Discussion: It is important for a surgeon to consider where the cement restrictor will sit in the femur during pre-operative templating in Total Hip Replacement. When the cement restrictor is going to be deployed beyond the femoral isthmus, an alternate method of cement restriction may need to be used. Universal sized plugs (e.g. Hardinge) function poorly in this situation. Press-fit plugs such as Biostop and Exeter have been previously shown to allow the generation of high pressures in bone cement when sited above the femoral isthmus or in stove pipe femurs. However their function is severely compromised when inserted past the femoral isthmus. Pre-injected cement plugs are variable in efficacy. The expandable Rex CementStop was simple to use and reliably occluded the femur, allowing the highest pressures to be generated.

Introduction: Aim of the study was to investigate the fixation patterns of cementless short-stem designs compared to the fixation patterns of well-known conventional stems artificially shortened in several steps.

Method: Primary rotational stability and tilting were measured in a standardized and validated simulator set-up. The tested short-stem designs included the Mayo (Zimmer), the CFP (Link), and the CUT (Eska). The “conventional” group consisted of the ABG (Stryker), the S-Rom (DePuy), and the G2 (DePuy). Additionally, these systems were tested when gradually shortened in several steps.

Results: The Mayo stem showed a 2/3 fixation pattern with 11.5 mdeg/Nm relative movement at the level of the lesser trochanter. The CFP showed a more distal fixation with 6.8 mdeg/Nm, and the CUT a proximal fixation with 7.6 mdeg/Nm. The Mayo and the CFP adapted to the femoral canal during ventro-dorsal tilting. The CUT tilted like a rigid body. This pattern of tilting was also shown by the ABG after it was shortened from 12.5 cm to 8.5 cm, by the G2 after a reduction from 15 to 6 cm, and the S-Rom after shortening from 18 to 6.9 cm.

Discussion: The shortened conventional stems showed a similar fixation pattern compared to the short-stem designs, the rotational stability was even better than with the Mayo and CFP stem. The CUT design acts like an extensively shortened conventional stem and seems to tilt in the femoral canal like a rigid body with a pressure increase at the lateral cortex of the femur.

Background: Metal-metal bearings are frequently implanted in young, active patients. The relationship between patient activity and Co and Cr ion levels has not been scientifically investigated.

Methods: Seven patient subjects with well-functioning metal-metal bearing hip prostheses and one control subject (no implants), all with normal renal function, were monitored during a two-week long activity protocol. Lower extremity activity was continuously assessed by a computerized, two-dimensional accelerometer (Step Activity Monitor; SAM). During the first week, subjects were requested to limit physical activity. Subjects then completed an hour-long treadmill test followed by a week where they were encouraged to be as physically active as possible. Serum Co and Cr ion levels and urine Cr levels were assessed at 10 different time points during these two weeks.

Results: Regardless of activity, the serum ion levels for a given patient were essentially constant and there was no correlation between patient activity and serum or urine ion levels. A mean increase in activity of 28% (95% CI, 13 to 43%; SE, 6%) during the high-intensity activity week resulted in a mean decrease of 2.6% (95% CI, −14.2 to 8.9%; SE, 4.7%) in serum Co and a mean increase of 2.0% (95% CI, −5.3 to 9.3%; SE, 3.0%) in serum Cr. During the treadmill test, a mean activity increase of 1,621% (16-fold) (95% CI, 972 to 2,271%; SE, 265%) resulted in a mean increase of 3.0% (95% CI, −5.7 to 11.7%; SE, 3.6%) in serum Co and a mean increase of 0.8% (95% CI, −3.5 to 5.0%; SE, 1.7%) in serum Cr. This effectively constitutes no change in serum ion levels for these changes in activity because the differences are within the variability for the measurement accuracy of these tests.

Conclusions: In patients with normal renal function and a well-functioning metal-metal bearing, ion levels are not affected by patient activity. Periodic measurements of serum ion levels can be used to monitor the tribologic performance of prosthesis with a metal-metal bearing without adjusting for patient activity.